Machine-Learning blends of geomorphic descriptors: value and limitations for flood hazard assessment across large floodplains

Abstract. Recent literature shows several examples of simplified approaches that perform flood hazard (FH) assessment and mapping across large geographical areas on the basis of fast-computing geomorphic descriptors. These approaches may consider a single index (univariate) or use a set of indices simultaneously (multivariate). What is the potential and accuracy of multivariate approaches relative to univariate ones? Can we effectively use these methods for extrapolation purposes, i.e. FH assessment outside the region used for setting up the model? Our study addresses these open problems by considering two separate issues: (1) mapping flood-prone areas, and (2) predicting the expected water depth for a given inundation scenario. We blend seven geomorphic descriptors through Decision Tree models trained on target FH maps, referring to a large study area (≈105 km2). We discuss the potential of multivariate approaches relative to the performance of a selected univariate model and on the basis of multiple extrapolation experiments, where models are tested outside their training region. Our results show that multivariate approaches may (a) significantly enhance flood-prone area delineation (overall accuracy: 93 %) relative to univariate ones (overall accuracy: 84 %), (b) provide accurate predictions of expected inundation depths (determination coefficient ≈0.7), and (c) produce encouraging results in extrapolation.

Download Full-text

Flood-prone area mapping using machine learning techniques: a case study of Quang Binh province, Vietnam

Natural Hazards ◽

10.1007/s11069-021-04821-7 ◽

2021 ◽

Author(s):

Chinh Luu ◽

Quynh Duy Bui ◽

Romulus Costache ◽

Luan Thanh Nguyen ◽

Thu Thuy Nguyen ◽

...

Keyword(s):

Machine Learning ◽

Machine Learning Techniques ◽

Learning Techniques ◽

Prone Area ◽

Flood Prone Area

Download Full-text

Combination of geomorphic classifiers through Machine Learning-based techniques for flood hazard assessment

10.5194/egusphere-egu21-4723 ◽

2021 ◽

Author(s):

Andrea Magnini ◽

Michele Lombardi ◽

Simone Persiano ◽

Antonio Tirri ◽

Francesco Lo Conti ◽

...

Keyword(s):

Machine Learning ◽

Hazard Assessment ◽

Water Depth ◽

Flood Hazard ◽

Research Centre ◽

Hazard Mapping ◽

Hazard Maps ◽

Flood Hazard Assessment ◽

Flood Prone Areas ◽

Flood Hazard Maps

Every year flood events cause worldwide vast economic losses, as well as heavy social and environmental impacts, which have been steadily increasing for the last five decades due to the complex interaction between climate change and anthropogenic pressure (i.e.&#160;land-use and land-cover modifications). As a result, the body of literature on flood risk assessment is constantly and rapidly expanding, aiming at developing faster, computationally lighter and more efficient methods relative to the traditional and resource-intensive hydrodynamic numerical models. Recent and reliable fast-processing techniques for flood hazard assessment and mapping consider binary geomorphic classifiers retrieved from the analysis of Digital Elevation Models (DEMs). These procedures (termed herein &#8220;DEM-based methods&#8221;) produce binary maps distinguishing between floodable and non-floodable areas based on the comparison between the local value of the considered geomorphic classifier and a threshold, which in turn is calibrated against existing flood hazard maps. Previous studies have shown the reliability of DEM-based methods using a single binary classifier, they also highlighted that different classifiers are associated with different performance, depending on the geomorphological, climatic and hydrological characteristics of the study area. The present study maps flood-prone areas and predicts water depth associated with a given non-exceedance probability by combining several geomorphic classifiers and terrain features through regression trees and random forests. We focus on Northern Italy (c.a. 100000 km2, including Po, Adige, Brenta, Bacchiglione and Reno watersheds), and we consider the recently compiled MERIT (Multi-Error Removed Improved-Terrain) DEM, with 3sec-resolution (~90m at the Equator). We select the flood hazard maps provided by (i) the Italian Institute for Environmental Protection and Research (ISPRA), and (ii) the Joint Research Centre (JRC) of the European Commission as reference maps. Our findings (a) confirm the usefulness of machine learning techniques for improving univariate DEM-based flood hazard mapping, (b) enable a discussion on potential and limitations of the approach and (c) suggest promising pathways for further exploring DEM-based approaches for predicting a likely water depth distribution with flood-prone areas.&#160;

Download Full-text

SISTEM INFORMASI GEOGRAFIS (SIG) DAERAH RAWAN BANJIR DI KOTA BENGKULU MENGGUNAKAN ARCVIEW

JURNAL MEDIA INFOTAMA ◽

10.37676/jmi.v12i1.271 ◽

2016 ◽

Vol 12 (1) ◽

Author(s):

Rahma Wayan Lestari ◽

Indra Kanedi ◽

Yode Arliando

Keyword(s):

Information System ◽

Geographic Information System ◽

Geographic Information ◽

The Body ◽

Microsoft Excel ◽

Microsoft Word ◽

Prone Area ◽

Flood Prone Area ◽

The City ◽

Flood Prone Areas

The purpose of this research is to create a geographic information system Bengkulu city flood-prone areas using ArcView. Apply the knowledge obtained during the lecture, especially relating to the development of Geographic Information Systems. To be able to produce a system that is accurate and useful information for the community. Where the research was conducted in the city of Bengkulu BASARNAS. Bengkulu BASARNAS office specializing in Search and Rescue (SAR), is the body that manage the flood of data that is still done manually, using Microsoft Word and Microsoft Excel. Thus experiencing problems in delivering information directly to the office because the SAR agencies require a long time.Keywords: Geographic Information System, Flood Prone Area

Download Full-text

Pemetaan Tingkat Kerawananan Banjir Di Kota Jambi Menggunakan Metode Scoring Dan Overlay Berbasis Sistem Informasi Geografis

Jurnal Ilmu dan Inovasi Fisika ◽

10.24198/jiif.v5i2.32860 ◽

2021 ◽

Vol 5 (2) ◽

pp. 132-141

Author(s):

Lusiani Pryastuti ◽

Keyword(s):

Land Use ◽

Information System ◽

Spatial Distribution ◽

Geographic Information System ◽

Soil Type ◽

Flood Vulnerability ◽

Overlay Method ◽

Prone Area ◽

Flood Prone Area ◽

Flood Prone Areas

This research is about flood vulnerability mapping in Jambi City based on Geographic Information System (GIS). This study is aiming to find out the flood vulnerability level, spatial distribution of flood, and flood prone areas in Jambi City. We used five parameters that affect flood vulnerability, including land slope, land level, land use, soil type, and rainfall during 2019. The method used is the scoring and overlay method with the help of ArcGis software. Flood vulnerability level was divided into three categories, namely quite vulnerable, vulnerable, and very vulnerable. The results obtained in this study are that most of Jambi City has a level of flood vulnerability in the vulnerable category, which is an area of 9254.82 ha (58%), while for the area that is dominated quite safe from flooding, Jambi Selatan sub-district, is 2849.14 ha (18%). This shows that more than half of the Jambi city area is a flood-prone area so it is very important to carry out structural and non-structural mitigation actions

Download Full-text

Analysis of the exposure of Polish lakes to flood risk

Limnological Review ◽

10.2478/limre-2015-0005 ◽

2015 ◽

Vol 15 (1) ◽

pp. 39-50

Author(s):

Anna Pasiecznik-Dominiak ◽

Andrzej Tiukało ◽

Grzegorz Dumieński

Keyword(s):

Flood Risk ◽

Flood Hazard ◽

Risk Level ◽

Resilience Theory ◽

Risk Area ◽

The Social ◽

Prone Area ◽

Extreme Hazards ◽

Flood Prone Area

Abstract Flooding constitutes one of the main natural hazards in Poland, which causes enormous social, economic and environmental losses. The main causes of the occurrence of floods include intensive rainfall, rapid melting of snow and ice cover, as well as strong gusts of wind from the sea. Based on the resilience theory (resistance, elasticity), which constitutes an efficient tool for the description of the social-ecological system capability or components thereof to mitigate the effects of dangerous events, as well as the capability of reconstructing and adapting the system to new conditions, the authors have analysed the exposure of Polish lakes to flood risks with a probability of occurrence Q0.2%, Q1% and Q10%. In order to determine the level of exposure of lakes to the risk of flooding by flood waters, studies were conducted using the flood hazard and flood risk maps which were developed under the Project entitled “IT System of the Country’s Protection against Extreme Hazards”. The result of the efforts of the group of authors is the determination of the number of lakes, which are located in the flood risk area Q0.2%, Q1% and Q10%, including division into risk level groups (low, moderate and high). The results presented in the paper may constitute a contribution to further, more detailed studies concerning assessment of the vulnerability of Polish lakes located in the flood prone area.

Download Full-text

Efficient pan-European river flood hazard modelling through a combination of statistical and physical models

10.5194/nhess-2017-4 ◽

2017 ◽

Cited By ~ 1

Author(s):

Dominik Paprotny ◽

Oswaldo Morales-Nápoles ◽

Sebastiaan N. Jonkman

Keyword(s):

Return Period ◽

Flood Hazard ◽

Physical Models ◽

Geographical Information ◽

Computational Time ◽

Joint Research ◽

Simplified Approach ◽

The Future ◽

Prone Area ◽

Flood Prone Area

Abstract. Flood hazard is being analysed with ever-more complex models on national, continental and global scales. In this paper we investigate an alternative, simplified approach, which combines statistical and physical models in order to carry out flood mapping for Europe. Estimates of extreme river discharges made using a Bayesian Network-based model from a previous study are employed instead of rainfall-runoff models. Those data provide flood scenarios for simulation of water flow in European rivers with a catchment area above 100 km2. The simulations are performed using a one-dimensional steady-state hydraulic model and the results are post-processed using geographical information system (GIS) software in order to derive flood zones. This approach is validated by comparison with Joint Research Centre's (JRC) pan-European map and five local flood studies from different countries. Overall, both our and JRC's maps have similar performance in recreating flood zones of local maps. The simplified approach achieved similar level of accuracy, while substantially reducing the computational time. The paper also presents the summarized results from the flood hazard maps, including future projections. We find that relatively small changes in flood hazard are observed (increase of flood zones area by 2–4 %). However, when current flood protection standards are taken into account, there is a sharp increase in flood-prone area in the future (28–38 % for a 1000 year return period). This is because in many parts of Europe river discharge with the same return period is projected to increase in the future, thus making the protection standards insufficient.

Download Full-text

GIS Based Method for Flood Hazard Assesment in Kobe River Watershed - North Maluku Province

E3S Web of Conferences ◽

10.1051/e3sconf/202132804019 ◽

2021 ◽

Vol 328 ◽

pp. 04019

Author(s):

Nani Nagu ◽

A. Latif Lita ◽

H Bebi ◽

Nurhalis Wahiddin

Keyword(s):

Flood Hazard ◽

Vulnerability Index ◽

Hazard Map ◽

Flood Vulnerability ◽

Flood Susceptibility ◽

Flood Hazard Map ◽

Prone Area ◽

Flood Prone Area ◽

Hierarchy Process ◽

Selection Of

The objectives of this study are to mapping the hazard-prone area and to analyse the flood vulnerability index in Kobe Watershed, Central Halmahera District. In order to determine the optimal selection of weights for the factors that contribute to flood risk, GIS and multi-criteria decision analysis (MCDA) were used in conjunction with the application of the analytical hierarchy process (AHP) method to create the flood hazard map. The flood hazard map was generated by using selected hazard factors including land use, topography, slope, and rainfall pattern. The result shows that the Kobe River basin is a flood-prone area, with 77.46 percent of its land classified as less prone to flooding and 21.41 percent classified as flood-prone. However, only 21.41 percent of its land is classified as flood-prone. Only 1.13 percent of the land is protected from the danger of floods, compared to the whole country. The altitude factor is the most important element influencing flood susceptibility in Weda District, where the majority of the land (16.34 percent) is located at or below sea level, making it particularly vulnerable to flooding.

Download Full-text

PEMETAAN BAHAYA BANJIR LAHAN SAWAH DI KABUPATEN KARAWANG

Jurnal Ilmu Tanah dan Lingkungan ◽

10.29244/jitl.19.1.41-45 ◽

2019 ◽

Vol 19 (1) ◽

pp. 41-45

Author(s):

Tommi Tommi ◽

Baba Barus ◽

Arya Hadi Dharmawan

Keyword(s):

Paddy Field ◽

Flood Hazard ◽

The Other ◽

Flood Mapping ◽

Flood Events ◽

Soil Drainage ◽

Prone Area ◽

High Flood ◽

Flood Prone Area ◽

Hazard Level

Flooding is one of the natural disasters that frequently hit several countries, including Indonesia. Data from the BNPB show of the year 1815 - 2013 ranks first flood disaster events most of the other disasters that as many as 5,394 events. Karawang District was ranked 3rd highest number of flood events in West Java. Nationally data from BNPB show Karawang ranks 8th flood-prone area. The purpose of this study to analyze the level of hazard of flooding the paddy field in Karawang. The method used in analyzing the level of hazard of flooding is done by overlaying and scoring from the paddy fields map, rainfall maps, soil drainage maps, and flood events maps. The results of this study indicate the paddy field in Karawang District which has a high flood hazard level contained in the Telukjambe West, East Telukjambe, and Jayakerta Sub Disctrict. Keywords: Hazard, flood, mapping

Download Full-text

Identifying the essential flood conditioning factors for flood prone area mapping using machine learning techniques

CATENA ◽

10.1016/j.catena.2018.12.011 ◽

2019 ◽

Vol 175 ◽

pp. 174-192 ◽

Cited By ~ 32

Author(s):

Mahyat Shafapour Tehrany ◽

Simon Jones ◽

Farzin Shabani

Keyword(s):

Machine Learning ◽

Machine Learning Techniques ◽

Conditioning Factors ◽

Learning Techniques ◽

Prone Area ◽

Flood Prone Area

Download Full-text

Mapping of Flooded Areas in the Kudus District

Indonesian Journal of Electrical Engineering and Computer Science ◽

10.11591/ijeecs.v4.i3.pp670-677 ◽

2016 ◽

Vol 4 (3) ◽

pp. 670

Author(s):

Rina Fiati ◽

Anastasya Latubessy

Keyword(s):

Information System ◽

Expert System ◽

Research Use ◽

Solid Objects ◽

Prone Area ◽

Rule Method ◽

Flood Prone Area ◽

Flood Flow ◽

Flooded Areas ◽

Flood Prone Areas

Flood is still an annual problem in the Kudus District. Based on the survey and interview with Regional Disaster Management Agency (BPBD – Badan Penanggulangan Bencana Daerah) data showed that in the Kudus District there are still many flood-prone areas. They also said that, there are six parameter that can be used to identify potential flood area such as: extensive inundation (km2, ha), depth or height of flood waters (meters), the flow velocity (m/s,km/h), the material washed away by flood flow (rocks, boulders, trees, and other solid objects), concentrations of water or silt thickness (meters, centimeters), and duration of inundation (hours, days, months). Therefor this research use six parameters are then analyzed and used as a benchmark model to identify flood-prone areas by using the production rule method, and as the material in constructing and designing flood-prone area identification systems based on expert system. Thus this research resulted a system to assist the identification of flood prone areas in the Kudus District by using expert system and geographic information system (GIS).

Download Full-text